Validation of recombinant Sendai virus in a non-natural
U Griesenbach1,4,5, G McLachlan2,4,5, T Owaki3, L Somerton1,4, T Shu3, A Baker2,4, P Tennant2,4, C Gordon2,4,
C Vrettou2,4, E Baker2,4, DDS Collie2,4, M Hasegawa3and EWFW Alton1,4
We have previously shown that recombinant Sendai virus (SeV) vector, derived from murine parainfluenza virus, is one of the
most efficient vectors for airway gene transfer. We have also shown that SeV-mediated transfection on second administration,
although reduced by 60% when compared with levels achieved after a single dose, is still high because of the efficient
transfection achieved by SeV vector in murine airways. Here, we show that these levels further decrease on subsequent doses.
In addition, we validated SeV vector repeat administration in a non-natural host model, the sheep. As part of these studies
we first assessed viral stability in a Pari LC Plus nebuliser, a polyethylene catheter (PEC) and the Trudell AeroProbe. We also
compared the distribution of gene expression after PEC and Trudell AeroProbe administration and quantified virus shedding after
sheep transduction. In addition, we show that bronchial brushings and biopsies, collected in anaesthetized sheep, can be used
to assess SeV-mediated gene expression over time. Similar to mice, gene expression in sheep was transient and had returned
to baseline values by day 14. In conclusion, the SeV vector should be strongly considered for lung-related applications
requiring a single administration of the vector even though it might not be suitable for diseases requiring repeat administration.
Gene Therapy (2011) 18, 182–188; doi:10.1038/gt.2010.131; published online 21 October 2010
Keywords: Sendai virus; cystic fibrosis; lung; gene transfer; sheep
Sendai virus (SeV), a murine paramyxovirus, is one of the most
efficient viral vectors for airway gene transfer.1,2SeV carrying the
cystic fibrosis (CF) transmembrane conductance regulator cDNA is
able partially to correct the characteristic CF transmembrane con-
ductance regulator-dependent chloride transport defect in the nasal
epithelium of CF knockout mice.3Several aspects of SeV biology may
explain the high gene transfer efficiency into airway epithelial cells.
Moreover, SeV uses cholesterol and sialic acid as receptors and both
are present on the surface of most cell types. Further, SeV requires
short contact time with the target cell for internalization, and
replicates in the cytoplasm of transduced cells, circumventing the
nuclear membrane barrier. In mice gene expression is transient,
with peak expression approximately 48h after transfection, generally
returning to baseline values within 2 weeks of transfection.1
A transmission-incompetent SeV vector has been developed by
deleting the F-protein, which is essential for cellular entry of the
viral genome (DF/SeV).4Moreover, this modification does not reduce
transfection efficiency of the virus.1,4
In general the level of transgene expression achieved from repeat
delivery of a viral vector is greatly reduced when compared with that
from a single dose due to induction of effective immune responses in
the recipient.5,6We have previously shown that SeV-mediated gene
expression is reduced by 60% on second dose, and that tolerization of
mice against immune-dominant SeV epitopes does not improve
efficacy.7However, given the extremely high transfection efficiency
of SeV, the levels of gene expression achieved after repeat administra-
tion may still be of sufficient therapeutic value, if retained on
subsequent administrations. Here, we assessed SeV-mediated trans-
fection efficiency after three doses of the virus and compared
residual levels of gene expression with that achieved using plasmid
DNA complexed to the cationic lipid GL67A, currently being
used in a clinical trial by the UK CF Gene Therapy Consortium
The vast majority of repeat administration studies have been
performed in mouse inbred strains, such as Balb/C and C57BL/6.
Given that SeV has a natural tropism for the murine lung leading to
pneumonia, we were concerned that assessment of repeat adminis-
tration in mice may not be representative of non-natural host models,
including man. The greater degree of similarity in size, structure and
physiology between the sheep and human lung led us to develop sheep
as an intermediate model for airway gene transfer,8to bridge the gap
between studies in rodents and the clinic. Here, we have also assessed
the transduction efficiency of both single and repeat SeV vector
administration in the ovine lung.
Levels of gene expression after repeat administration of DF/SeV
with short-term intervals to mouse lung are similar to non-viral
We have previously shown that SeV-mediated transfection on second
administration, although reduced by 60% when compared with levels
Received 19 May 2010; revised 27 July 2010; accepted 9 August 2010; published online 21 October 2010
1Department of Gene Therapy, Imperial College London, National Heart and Lung Institute, London, UK;2The Wellcome Trust Centre for Research in Comparative Respiratory
Medicine, The Roslin Institute & R(D)SVS, University of Edinburgh, UK;3DNAVEC Corporation, Tsukuba, Japan and4UK CF Gene Therapy Consortium, UK
Correspondence: Dr U Griesenbach, Department of Gene Therapy, Imperial College at the National Heart and Lung Institute, Manresa Road, London SW3 6LR, UK.
5These authors contributed equally to this work.
Gene Therapy (2011) 18, 182–188
& 2011 Macmillan Publishers Limited All rights reserved 0969-7128/11
achieved after a single dose,7is still high because of the efficient
transfection achieved by SeV vector in murine airways. Here, we
assessed whether these residual levels are maintained on subsequent
dosing. Murine lungs were transfected with three doses (108cell
infectious unit (CIU)/mouse/dose) of transmission-incompetent
DF/SeV at either weekly or monthly intervals by nasal sniffing.
To circumvent generation of anti-reporter gene immune responses,
a vector without a reporter gene (DF/SeV-empty) was administered
for the first two doses. A vector carrying a luciferase reporter gene
(DF/SeV-luc) was administered as the third dose, and luciferase (luc)
expression was quantified 2 days after administration (n¼16–19/
group). Luc expression was compared with mice receiving only a
single dose of virus (n¼20), mice transfected with the non-viral
vector GL67A complexed to pCIK-Lux (n¼19) and phosphate-
buffered saline (PBS)-treated negative control mice (n¼10). Gene
expression after 3 weekly or monthly, doses was significantly
(Po0.001) reduced by 3 to 4 logs compared with mice receiving
only one dose, but was significantly (Po0.05) higher than the negative
controls (Figure 1). Residual expression levels after three doses
of DF/SeV were not significantly different to non-viral transfection
SeV viability in the Trudell AeroProbe catheters is suitable for
in vivo administrations
We next assessed transfection efficiency and repeat administration
in sheep for which the virus does not have a natural tropism. In
preparation for the in vivo sheep experiments, we first assessed DF/SeV
viability in a Pari LC Plus nebuliser, which we routinely use for non-
viral gene transfer in sheep. We assessed viability by either collecting
DF/SeVaerosol at the end of the Pari LC Plus mouth-piece (as used in
man) or at the end of an endotracheal tube (as used for sheep
nebulization studies). DF/SeV viability was reduced to less than 1%
in the ‘mouth-piece’ set-up and dropped to less than 0.01% when
connected to an endotracheal tube (three independent experiments).
This delivery method was, therefore, not used in any further study.
We next assessed virus viability in a polyethylene catheter (PEC),
which can be fed through the biopsy channel of a bronchoscope and
used to administer virus as a bolus into specific lung segments, and a
Trudell AeroProbe catheter, which can be positioned within the
trachea or bronchi and generates an aerosol at the tip. Virus viability
was 100 and 49±3% in the PEC and AeroProbe, respectively (three
PEC and AeroProbe administration of DF/SeV-LacZ leads to
wide-spread b-gal expression in the sheep lung but the distribution
DF/SeV-LacZ (3.4?109CIU/sheep. in 5ml, n¼3 sheep) was first
administered by PEC as a bolus into a single lung segment. Gene
expression was visualized with X-gal staining 48h after transfection.
High level, but patchy, transfection was seen in the lung possibly
due to uneven vector distribution and pooling of the bolus volume
compatible with catheter-based delivery (Figure 2a). Using segmental
instillation the adjacent lung segments showed no evidence of
transgene expression (Figure 2b).
We next administered DF/SeV-LacZ using the Trudell AeroProbe to
either a single lung segment (1010CIU per sheep in 5ml, n¼1) or the
whole lung (4?1010CIU per sheep in 24ml, n¼1). When directed into
a single segment we observed spill over to the adjacent segment on the
same (Figures 2c and d), but not to the contralateral side, of the lung
(data not shown). This is likely due to a combination of pooling of
the aerosol following impaction on the airway walls close to the tip of
the catheter and the fact that the bronchoscope is not wedged in the
airway. Figures 2e and f demonstrates deposition of aerosol on airway
bifurcations and, interestingly, within a sub-mucosal gland and duct.
With the Trudell AeroProbe positioned in the trachea we observed
more even, wide-spread distribution throughout the lung and less of
a pooling effect, likely due to the AeroProbe tip not being so close
to airway walls (Figure 2g). The Trudell AeroProbe was, therefore,
used for subsequent experiments.
Infectious virus shedding does not occur after DF/SeV transfection
Infectious virus shedding after sheep transfection would be a concern,
because both the environment and the operators would require
protection from exposure to virus. We, therefore, determined whether
virus shedding had occurred. Nasal and oral swabs, bronchoalveolar
lavage fluid and lung tissues were collected 48h after transfection with
DF/SeV-LacZ (1010CIU in 24ml per sheep, n¼5 sheep). Three plates
were analyzed per sample per sheep. Virus spiked positive controls
and untreated samples were analyzed in parallel. A small number of
Xgal-positive cells (one to three cells per plate) were detected in the
collected fluids and tissues of transfected sheep, but these were also
present in the untreated controls. In a second group of animals (n¼3)
sampled in vivo 48h post-administration we saw no X-gal-positive
cells from oral or nasal swabs, broncho-alveolar lavage fluid or serum
samples. We conclude that infectious virus shedding does not occur at
a detectable level at 48h post-transfection.
Bronchial brushings and biopsies can be used to assess
Although LacZ is a useful reporter gene to visualize gene expression
histologically, we were not certain whether b-gal expression would be
Figure 1 Repeat administration of DF/SeV in mouse lung. Murine lungs were
transfectedwith threedoses (108CIU/mouse/dose)
incompetent DF/SeV at either weekly or monthly intervals by nasal sniffing.
To circumvent generation of anti-reporter gene immune responses a vector
without a reporter gene (DF/SeV-empty) was administered for the first two
doses. A vector carrying a luciferase reporter gene (DF/SeV-luc) was
administered as the third dose and luciferase (luc) expression in lung was
quantified two days after administration (n¼16–19 mice/group). Luc
expression was compared with mice receiving only a single dose of virus
(n¼20), mice transfected with the non-viral vector GL67A complexed to
pCIK-Lux (n¼19) and PBS-treated negative controls (n¼10). Each symbol
representsonemouse. The horizontal
***Po0.001 compared with single-dose cohort.#Po0.05 compared with
barindicates group mean.
Validation of recombinant Sendai virus in sheep
U Griesenbach et al
quantifiable in bronchial brushings and biopsies, used to assess
duration of gene expression in vivo. We, therefore, collected bronchial
brushings (n¼4/sheep) and biopsies (n¼4/sheep) before transfection,
then administered DF/SeV-LacZ (1010CIU in 24ml/sheep, n¼3 sheep)
and collected bronchial brushings and biopsies (n¼4 each for one
sheep, n¼8 each for the other two sheep) and tissue pieces at necropsy
48h post transfection. Using a chemiluminescent b-gal assay we were
able to detect robust levels of b-gal in cell lysates prepared from
Figure 2 b-gal expression in sheep lung 2 days after transfection with DF/SeV-LacZ delivered with either a polyethylene catheter (PEC) or the Trudell
AeroProbe. DF/SeV-LacZ was administered to a single segment of the sheep lung with a PEC (3.4?109CIU/sheep in 5ml, n¼3 sheep). High level, but
patchy, gene expression was seen in the treated segment (a), but not in the adjacent segment (b). DF/SeV-LacZ was administered to a single segment using
the Trudell AeroProbe (1010CIU/sheep in 24ml, n¼1). Gene expression was visible in the treated, as well in the adjacent segment (c, d). In addition b-gal
expression was detected on airway bifurcations and within a sub-mucosal glands (e, f). DF/SeV-LacZ was administered to a whole lung using the Trudell
AeroProbe (4?1010CIU/sheep in 24ml, n¼1) and wide-spread even distribution throughout the lung was observed (g). Representative images are shown.
Validation of recombinant Sendai virus in sheep
U Griesenbach et al
brushings and biopsies (brushings, pre: 66±25, post: 7781±6376
relative light unit per mg protein, Po0.01, biopsies, pre: 84±40, post:
3074±1551 relative light unit per mg protein, Po0.001). Levels of
gene expression in biopsies and brushings taken from the same lobe
correlated well (r2¼0.65, P¼0.003). However, we were unable to
reliably quantify b-gal expression in cell lysates prepared from tissue
pieces, most likely due to the intensive red colour of the lysate
interfering with the assay (data not shown). Thus, repeated sampling
of bronchial brushings and biopsies should allow repeated analysis
of gene expression over time in vivo.
DF/SeV gene expression in sheep is transient
We next administered a dose of DF/SeV-LacZ to each side of the lung
with the AeroProbe positioned in the right and left main bronchi.
Each side received a dose of 5?109CIU in 7ml (n¼4 sheep) and
bronchial biopsies were collected 2, 14, 28 and 56 days after transfec-
tion (n¼8/sheep/time point) to follow gene expression over time by
comparisonwith pre-treatment biopsies (n¼6/sheep). Similar to mice,
the expression was transient and had returned to baseline values by
day 14 (Figure 3).
Reduced expression from repeat administration of DF/SeV in sheep
To determine the efficacy of repeat administration, sheep were
transfected with three doses (1010CIU/sheep/dose) of DF/SeV at
monthly intervals. To circumvent generation of anti-reporter gene
immune responses DF/SeV-empty was administered for the first two
doses. DF/SeV-LacZ was administered as the third dose and b-gal
expression was quantified in bronchial biopsies (n¼20/sheep) 2 days
after the final administration. A negative control group was treated
with three doses of DF/SeV-empty (n¼4 sheep/group). Although
we detected some residual expression in one sheep (Po0.05),
overall there was no evidence that re-administration of SeV into
sheep is feasible (Figure 4). We, therefore, concluded that monthly
re-administration of DF/SeV in sheep results in a similar loss of
efficiency as we observed in mice. For comparison, GL67-mediated
non-viral gene transfer does not lead to detectable levels of
reporter protein expression in sheep bronchial biopsies or brushings
(data not shown).
SeVand the modified transmission incompetent DF/SeVare one of the
most efficient gene transfer agents for airway epithelial cells, but
here we show that repeat administration (three doses at short-term
intervals) into mice, which is the natural host, is inefficient.
In addition, we optimized SeV-mediated gene transfer into sheep
lung for which the virus has no natural tropism, achieving efficient
transduction. However, as in mice, repeat administration was unable
to achieve the high levels of gene transfer that could be demonstrated
following a single dose.
Reports on repeat administration of adenovirus or adeno-associated
virus to the lungs vary, with some studies reporting successful
re-administration,9–11whereas other studies failed to detect gene
expression after repeat administration.6,12Differences are likely
due to different models being used, as well as different number of
doses being administered. In some publications, single-dose controls
are also missing, which makes an assessment of efficacy after repeated
The number of repeat administrations (two vs three) is an
important variable when performing and interpreting the efficacy of
repeat-dose virus administration. Several population vaccination
programmes, for example, human papillomavirus and hepatits
A vaccine, require a three-dose schedule. In the context of virus-
mediated gene transfer, it is feasible that the immune system may not
be fully activated after only two doses of virus. Consequently, analysis
of gene expression after additional doses is necessary before definite
conclusions about repeat administration of viral vectors can be
drawn. This is supported by our data. After a second dose of
DF/SeV, expression is significantly reduced by 60 and 490% in
mice7and sheep, respectively. However, because of the original
extremely high gene transfer efficiency the residual gene expression
after the second dose is still significantly higher than untransfected
controls, or a single dose of non-viral vector. After subsequent doses,
however, gene expression drops further leading to very low levels of
residual gene expression. In mice these levels are similar to GL67A-
Figure 3 Duration of gene expression in sheep lung. Sheep were transfected
with DF/SeV-LacZ (1010CIU in 24ml per sheep, n¼4 sheep) and biopsies
were collected 2, 14, 28 and 52 days after transfection (n¼8/sheep/time
point) to follow gene expression over time. Each symbol represents one
biopsy. Horizontal bars indicate group mean. Individual sheep are marked as
different symbols. ***Po0.001 compared with all other groups.
Figure 4 Repeat administration (3 doses monthly) of DF/SeV-LacZ in sheep.
Sheep lung was transfected with three doses (1010CIU/sheep/dose) of
DF/SeV at monthly intervals. To circumvent generation of anti-reporter gene
immune responses DF/SeV-empty was administered for the first two doses.
DF/SeV-LacZ was administered as the third dose (n¼4 sheep, open symbols)
and b-gal expression was quantified in bronchial biopsies (n¼20/sheep)
2 days after administration. A negative control group was treated with
three doses of DF/SeV-empty (n¼4 sheep, closed symbols). Each symbol
represents one biopsy (n¼19–20 biopsies/sheep). Horizontal bars indicate
group medians. Individual sheep is marked as different symbol. *Po0.05
compared with all other sheep.
Validation of recombinant Sendai virus in sheep
U Griesenbach et al
mediated expression and repeat administration of SeV does not,
therefore, offer an advantage over the cationic lipid GL67A. The latter
is the current ‘gold-standard’ lipid for airway transfection and is
currently used by us in clinical trials for CF gene therapy. Moreover,
repeat administration of DF/SeV into its natural host (mice) and
non-natural host (sheep) models led to similar results.
This study also assessed our understanding of DF/SeV vector
delivery to the lungs of large animal models. Thus, (1) we showed
that, DF/SeV is not stable in the commonly used Pari-LC Plus jet
nebuliser. Although this nebuliser is suitable for aerosolization of
lipid/DNA complexes13and non-enveloped adeno-associated virus14
and has been used in gene therapy clinical trials,15,16the shear forces
generated appear to destroy the enveloped SeV. The latest generation
of vibrating mesh-based single-pass nebulisers, such as the Pari eFlow,
will need to be assessed for the delivery of enveloped virus.
(2) DF/SeV was stable in polythylene catheters (100% stability) and
the Trudell AeroProbe (approximately 50% stability). The decrease in
viability observed with the Aeroprobe was not unexpected as only
40% of viable particles were recovered after aerosolization of helper-
dependent adenovirus vectors through catheters rated at 8mm, rising
to 80% with 15mm, respectively.17
(3) In addition, we were able to compare virus distribution after
catheter-based segmental administration and Trudell AeroProbe aero-
solization. Delivery by instillation results in patchy expression in the
lungs likely due to the uneven distribution and pooling effect from the
delivery of a bolus. The AeroProbe catheter has the advantage that
it generates an aerosol at the tip of the catheter. This results in a
cone-shaped particle stream of aerosol visible to the eye even when
viewed through the bronchoscope camera. To target an individual
segment, we positioned the Aeroprobe in a small bronchus (with an
approximate diameter of o5mm) leading to that segment and
observed that a significant proportion of the aerosol stream impacted
on the airway wall close to the catheter tip and collected in a puddle.
This resulted in a transgene expression pattern similar to that observed
with the instillation delivery and significant spill-over to the adjacent
In contrast, when the Aeroprobe was positioned either in the
trachea or the main bronchus (approximate diameter of 10mm),
the larger airway diameter reduced the impaction of the aerosol on the
airway wall and led to a more even distribution in the lung. The
predicted aerosol droplet size generated by this catheter is considerably
larger (5–10mm) than would normally be expected to result in
deposition in the terminal bronchioles or alveolar regions. However,
intratracheal aerosolization of similar size droplets to the rabbit lung,
or of droplets with a mass median aerodynamic diameter of 430mm
in the mouse, has been shown to give deposition to airways of all sizes
and even into the alveolar regions17,18suggesting that the droplet size
is not so critical when generated intratracheally.
(4) We did not detect any infectious virus shedding 48h after
transfection and were, therefore, able to house treated sheep in outside
pens without any specific precautions for operators and waste disposal
after this time-point. Before 48h post-dosing, animals were housed in
indoor pens and all waste was autoclaved consistent with guidelines
for working with genetically modified organisms.
(5) Administration of DF/SeV-Luc lead to gene expression 3.5 logs
above untransfected controls in bronchial biopsies (unpublished data,
DF/SeV-Luc: 7918±2634, untransfected: 1.55±0.37 RLUmg?1protein
n¼3), whereas DF/SeV-LacZ transfection only increased b-gal by 1 log
compared with control, when quantified with a chemiluminescent
reporter gene assay. In cell lysates prepared from lung tissue pieces
these differences were even more striking. DF/SeV-Luc administration
increased luc expression by 5-logs (unpublished data, DF/SeV-Luc:
21079±6659, untransfected: 0.07±0.006 RLUmg?1protein n¼3),
whereas b-gal was not reliably detectable after DF/SeV-LacZ transfec-
tion. For quantification of reporter gene expression in cell lysates
luciferase appears to be a more reliable and sensitive reporter gene in
sheep lung, consistent with results observed in murine lungs (unpub-
lished data). This is likely due to the red colour of the tissue lysate
interfering with the chemiluminescent b-gal assay used in this study.
SeV may not be the vector of choice for gene therapy for lung
diseases requiring longer-term expression because expression levels
after re-administration of the vector are low because of the induction
of potent cellular and humoral immune responses against the virus,
although the relative importance of humoral vs cellular immunity in
this process is unclear. In addition to neutralising antibodies, activa-
tion of cytotoxic T-cells natural killer (NK) cells are activated through
interaction of the SeV hemagluttin-neuraminidase (HN) protein with
the NK cell receptor NKp4619having an important role. The effect
of SeV on immune modulation is further highlighted in a study by
Komary et al, which showed that SeV-transduced dendritic cells can
induce persistent NK and CD4-cell-dependent anti-tumor activity.20
We have previously shown that SeVadministration evokes cellular and
humoral immune responses in an animal model and that the level of
neutralising antibodies (as measured by in vitro assays) increases after
virus re-administration.7However, there was no correlation between
neutralising antibody levels and residual gene expression (r2¼0.046,
P¼0.16). We have also shown that partial T-cell tolerance to SeV
infection can be induced in mice after administration of immuno-
dominant CD4 peptide eptiopes, but that a reduction in T-cells does
not alter the levels of SeV-neutralising antibodies, or allow for repeat
administration of the virus. Although in vitro quantification of
neutralising antibodies is frequently undertaken, we do not believe
that a single host defence factor will reliably predict efficiency of repeat
administration in pre-clinical studies. Our concern about quantifica-
tion of neutralising antibodies, in part, stems from recent studies
showing that (a) adeno-associated virus vector can be re-administered
even in the presence of high levels of circulating neutralising anti-
bodies21and (b) in vitro neutralization assays fail to predict inhibition
by antiviral antibodies in vivo.22
However, the vector remains one of the most efficient gene transfer
agents for the lung. The envelope proteins F (fusion) and HN
(hemagglutinin-neuraminidase) are the key factors in determining
the high-transfection efficiency to airways. We have, therefore, recently
pseudotyped a simian lentiviral vector with the SeV F and HN proteins
(F/HN-SIV)23and have shown that the virus efficiently transfects
mouse airway epithelium in vivo, as well as human air–liquid interface
cultures.24Moreover, F/HN-SIV-mediated gene expression persists for
more than 17 months after a single administration and repeated
administration (3 doses) is feasible.24It is interesting that the
HN-mediated activation of NK cells does not appear to interfere
with F/HN-SIV-mediated gene expression. The F/HN-SIV vector
may, therefore, be ideally suited for pulmonary gene therapy.
In conclusion, although SeV transduces the airway epithelium of
mice and sheep efficiently, and retains comparatively high levels of
gene expression after administration of a second dose, these levels
drop further following the administration of a third dose at weekly
or monthly intervals. Responses to repeat administration of SeV in
natural hosts (mice) and non-natural hosts (sheep) are, therefore,
similar. We suggest that SeV is not suitable for diseases that require
high-level gene expression after repeat administration, but should be
strongly considered for lung-related applications requiring a single
administration of the gene transfer vector.
Validation of recombinant Sendai virus in sheep
U Griesenbach et al
MATERIALS AND METHODS
The generation and propagation of the recombinant DF/SeV vector carrying a
luciferase (DF/SeV-luc) or LacZ (DF/SeV-LacZ) reporter gene or no reporter
gene (DF/SeV-empty) was carried out as previously described.25The super-
natant of LLC-MK2/F7 cells containing infectious particles was subsequently
purified, concentrated and stored at –801C. Virus titre was determined by
infecting LLC-MK2 cells and counting the number of b-gal-expressing cells
after X-gal staining or by using an anti-SeVantibody and fluorescent immuno-
histochemistry. The titre was expressed as CIUs perml.
Virus stability in delivery devices
PariLC plus. Two nebuliser set-ups were assessed in vitro. (a) Nebuliser with
mouth-piece, (b) Nebuliser with attached endotracheal tube used for aerosol
delivery to sheep. 5ml of DF/SeV-LacZ (107CIUml?1) were nebulized using a
pressure of 29 psi until run dry (approximately 15min). Aerosols were collected
in a plastic tube. Polythylene endoscopic wash catheter (PEC, Olympus):
5ml of DF/SeV-LacZ (107CIUml?1) were delivered through the catheter and
collected in a plastic tube. Trudell AeroProbe catheter (Trudell Medical
International, Ontario, Canada). This is a multi-lumen catheter, with liquid
injected down a central lumen and sheared into droplets at the distal tip by
high-pressure air travelling down six peripheral lumens. 5ml of DF/SeV-LacZ
(107CIUml?1) were delivered through the AeroProbe and collected in a plastic
tube. To quantify virus viability, confluent LLC-MK2 cells grown in six-well
plates were transfected in triplicate with appropriate virus dilutions
(100ml/plate). After 1h cells were washed once with PBS after which 2ml of
MEM+10 foetal bovine serum and 1% penicillin/streptomycin were added. X-
gal staining was performed 48h after transfection. X-gal-positive cells were
quantified in eachwell. Virus not exposed to nebulisers or catheters was used as
a positive control and untransfected cells were used as a negative control. Virus
titre (CIUml?1) was defined as X-gal positive per ml and % cell viability was
Preparation of GL67A/DNA complexes
A eukaryotic expression plasmid carrying the luciferase reporter gene cDNA
(pCIKLux) under the control of the human cytomegalovirus immediate early
promoter/enhancer (CMV) was used. Cationic lipid GL67A was supplied by
Genzyme Corporation (Framingham, MA, USA) and complexed with DNA as
previously described for intrapulmonary administration by nasal ‘sniffing’.26
In vivo gene transfer in mice and sheep
All experiments were carried out with approval of appropriate local Ethics
Committees and according to Home Office regulations.
(Medical Developments Australia, Springvale, Australia). For intrapulmonary
administration by nasal ‘sniffing’ a single 100ml bolus containing either virus
(108CIU/mouse or GL67A/DNA complexes (80mg plasmid DNA complexed to
GL67A at a 1:4 lipid:DNA molar ratio) was slowly pipetted onto the nose and
the solution was rapidly sniffed into the lungs. At indicated time-points the
administration was repeated, or animals were culled and lung tissue retrieved
for analysis of luciferase expression.
Female Balb/C mice (6–8 weeks) were anaesthetised with metofane
before the study began and underwent a preliminary examination involving
bronchoscopic visualization and bronchoalveolar lavage of segment right apical
under gaseous anaesthesia 1–3 weeks before treatment to confirm absence of
pre-existing pulmonary disease. Anaesthetized sheep were maintained in a
whole body, negative pressure respirator as described previously.8Virus was
delivered by means of a bronchoscope, either by (a) instillation to single lung
segments via the PEC, (b) by AeroProbe catheter directed at specific locations,
or (c) by AeroProbe catheter directed at the whole lung. For instillation to
individual lung segments the bronchoscope was wedged within that segment
and either 0.67?109CIUml?1of DF/SeV-LacZ (segment right caudal diaphrag-
matic) or empty vector (DF/SeV-empty (segment left caudal diaphragmatic)
Suffolk Cross ewes, 35–60kg, were treated with anthelminthic agents
instilled in 5ml PBS. For AeroProbe delivery to individual segments the tip
of the AeroProbe was positioned in the lumen at the entrance to the segment,
but not wedged, as 50-psi air pressure is required to generate the aerosol.
2?109CIUml?1were delivered in 5ml PBS.
For delivery to the whole lung the AeroProbe was positioned centrally
in the trachea, immediately distal to the end of the endotracheal tube
expression and repeat dose studies a 7ml dose of virus containing 5?109or
1?1010CIU was delivered to each side of the lungs, respectively. With the
Aeroprobe catheter positioned far back in the right and left main bronchi the
aerosol exposure was focussed in the caudal and ventral diaphragmatic lung
At the indicated time points the temperature of the animal was recorded and
blood samples collected for serum and haematology. The sheep were either
killed for analysis, anaesthetized as before for repeat administration of virus or
anaesthetized and maintained on a positive pressure ventilation system
(Harvard Apparatus Model 708) for the collection of bronchial biopsies and
bronchial brushings with biopsy forceps (Crocodile Biopsy Forceps, Olympus
FB-15K-1, Olympus Keymed, Essex, UK) or cytology brushes (Olympus BC-
202D-2010), respectively. Bronchial brushings were collected into PBS and
biopsies frozen for luciferase assays. Following euthanasia by lethal injection
and exsanguination, the lungs were removed for tissue harvesting. The
pulmonary circulation was flushed through the pulmonary artery with 2–3l
of PBS before sampling. bronchoalveolar lavage fluid was collected as for pre-
treatment sampling but from segment right caudal diaphragmatic. Lung tissues
collected post mortem were fixed by perfusion with 2% neutral buffered
formalin, 0.2% glutaraldehyde, 2mm MgCl2and 5mM EGTA pH 8 in 0.1 M
phosphate buffer (pH 7.3) and selected segments dissected out for X-gal
staining. Alternatively, individual segments were cut transversely into approxi-
mately 1-cm thick slices representing upper, middle and lower regions.
Individual airways (42mm diameter) were dissected from the upper region
(Airway Upper). Parenchymal samples (containing airways too small to dissect,
o2mm diameter) were derived from the upper (parenchyma upper), middle
(M) and lower (L) regions. Finely chopped samples were snap frozen for
deliveredin 24mlPBS. For duration of
Reporter gene expression
(Promega, Southampton, UK) and homogenized, followed by three freeze–
thaw cycles (?801C for 30min, thawed at room temperature) and centrifuga-
tion (10min at 13000gav). Sheep lung tissue (300mg) or whole-bronchial
biopsies were transferred to FastRNA ProGreen matrix tubes (MP Biomedicals,
Solon, OH, USA) in 600ml reporter lysis buffer (RLB; Promega Corp.,
Madison, WI, USA) and homogenized in a FastPrep Instrument (MP Bio-
medicals) for 40s at speed setting 6. Lysates were centrifuged at 13000 gav/
10min/41C. Supernatant was frozen on dry ice and stored at ?801C. All
supernatants were removed and frozen at ?801C for luciferase quantification.
Luciferase activity was measured in the supernatant using a standard luciferase
assay kit (Promega) and the TD-20e luminometer (Turner BioSystems,
Sunnyvale, CA, USA) or the Anthos Lucy1 luminometer (Labtech Interna-
tional, East Sussex, UK). b-gal assay in tissue lysate: sheep lung tissue was
homogenized (Ultra-Turrax homogeniser Science Lab Houston, TX, USA) in
1?RLB buffer by weight (1ml RLB to 1g tissue and bronchial biopsies were
homogenized by hand using scissors in 120ml 1?RLB buffer. Supernatants
from the homogenates were collected after three freeze–thaw cycles (?801C for
30min, thawed at room temperature) and centrifugation (10min at 13000gav).
Samples were stored –801C b-gal expression was quantified using the Clontech
Detection Kit II (BD Biosciences Clontech, Franklin Lakes, NJ, USA). Light
emission was measured by the TD-20e luminometer as described above. Protein
assay: total protein per sample was determined using the BioRad protein assay
kit (BioRad laboratories, Hercules, CA, USA) and luciferase or b-gal activity
was expressed as arbitrary relative light units per mg total protein. X-gal
staining: lung tissue was stained as described previously.27Briefly, fixed sheep
tissue was washed three times in detergent (20min each) and stained in X-gal
for up to 48h at 301C. Samples were then washed (2?20min in PBS
containing 2mM MgCl2) and then processed into wax for sectioning.
Mouse right lungs were placed in 300ml 1?RLB buffer
Validation of recombinant Sendai virus in sheep
U Griesenbach et al
Virus shedding Download full-text
Trudell AeroProbe administration (1010CIU in 24ml PBS). At 48h after
delivery bronchoalveolar lavage, oral and nasal swabs in PBS and lung tissue
samples were collected from two sheep at necropsy. For the remaining three
sheep nasal and oral swabs, broncho-alveolar lavage fluid and serum were
collected from sheep that were kept alive as part of the 28-day duration study.
These samples were analyzed for viable virus particles. Lung tissue was
homogenized in PBS and all samples were filtered (0.45mm) to allow removal
of bacteria and fungi. Moreover, filters were first treated with 1ml of DF/SeV-
GFP (107CIUml?1) to minimise DF/SeV-LacZ absorption to the filter during
this process. Viable virus particles were then quantified by incubating the
samples with LLC-MK2 cells as described above.
To determine assay sensitivity, PBS was ‘spiked’ with known amounts of
virus (7 to 7000CIUml?1), filtered through a 0.45mm filter and plated onto
LLCMK cells for X-gal staining. The assay was able to detect 1:250 viable virus
Statistical analyses were performed by ANOVA or Kruskal–Wallis followed by
post-hoc analysis appropriate for parametric and non-parametric data. The null
hypothesis was rejected at Po0.05.
CONFLICT OF INTEREST
MH and TS are members of the corporate management of DNAVEC
Corporation. The remaining authors declare no conflict of interest.
We thank Dr M Inoue (DNAVEC, Corporation) for help with preparing the
manuscript. This work was in part funded by the Cystic Fibrosis Trust and
a CF Trust Senior Fellowship (UG). The work was supported by the NIHR
Respiratory Disease Biomedical Research Unit at the Royal Brompton and
Harefield NHS Foundation Trust and Imperial College London.
1 Ferrari S, Griesenbach U, Shiraki-Iida T, Shu T, Hironaka T, Hou X et al. A defective
nontransmissible recombinant Sendai virus mediates efficient gene transfer to airway
epithelium in vivo. Gene Ther 2004; 11: 1659–1664.
Yonemitsu Y, Kitson C, Ferrari S, Farley R, Griesenbach U, Judd D et al. Efficient gene
transfer to airway epithelium using recombinant Sendai virus. Nat Biotechnol 2000;
Ferrari S, Griesenbach U, Iida A, Farley R, Wright AM, Zhu J et al. Sendai
virus-mediated CFTR gene transfer to the airway epithelium. Gene Ther 2007; 14:
Li HO, Zhu YF, Asakawa M, Kuma H, Hirata T, Ueda Y et al. A cytoplasmic RNA vector
derived from nontransmissible sendai virus with efficient gene transfer and expression.
J Virol 2000; 74: 6564–6569.
Halbert CL, Rutledge EA, Allen JM, Russell DW, Miller AD. Repeat transduction in the
mouse lung by using adeno-associated virus vectors with different serotypes. J Virol
2000; 74: 1524–1532.
Sumner-Jones SG, Gill DR, Hyde SC. Lack of repeat transduction by recombinant
adeno-associated virus type 5/5 vectors in the mouse airway. J Virol 2007; 81:
7 Griesenbach U, Boyton RJ, Somerton L, Garcia SE, Ferrari S, Owaki T et al. Effect of
tolerance induction to immunodominant T-cell epitopes of Sendai virus on gene
expression following repeat administration to lung. Gene Ther 2006; 13: 449–456.
McLachlan G, Baker A, Tennant P, Gordon C, Vrettou C, Renwick L et al. Optimizing
aerosol gene delivery and expression in the ovine lung. Mol Ther 2007; 15: 348–354.
Auricchio A, O’Connor E, Weiner D, Gao GP, Hildinger M, Wang L et al. Noninvasive
gene transfer to the lung for systemic delivery of therapeutic proteins. J Clin Invest
2002; 110: 499–504.
10 Fischer AC, Beck SE, Smith CI, Laube BL, Askin FB, Guggino SE et al. Successful
transgene expression with serial doses of aerosolized rAAV2 vectors in rhesus
macaques. Mol Ther 2003; 8: 918–926.
11 Koehler DR, Martin B, Corey M, Palmer D, Ng P, Tanswell AK et al. Readministration of
helper-dependent adenovirus to mouse lung. Gene Ther 2006; 13: 773–780.
12 Harvey BG, Leopold PL, Hackett NR, Grasso TM, Williams PM, Tucker AL et al.
Airway epithelial CFTR mRNA expression in cystic fibrosis patients after repetitive
administration of a recombinant adenovirus. J Clin Invest 1999; 104: 1245–1255.
13 Eastman SJ, Tousignant JD, Lukason MJ, Murray H, Siegel CS, Constantino P et al.
Optimization of formulations and conditions for the aerosol delivery of functional
cationic lipid:DNA complexes. Hum Gene Ther 1997; 8: 313–322.
14 Leung K, Louca E, Munson K, Dutzar B, Anklesaria P, Coates AL. Calculating expected
lung deposition of aerosolized administration of AAV vector in human clinical studies.
J Gene Med 2007; 9: 10–21.
15 Alton EW, Stern M, Farley R, Jaffe A, Chadwick SL, Phillips J et al. Cationic
lipid-mediated CFTR gene transfer to the lungs and nose of patients with cystic fibrosis:
a double-blind placebo-controlled trial. Lancet 1999; 353: 947–954.
16 Moss RB, Rodman D, Spencer LT, Aitken ML, Zeitlin PL, Waltz D et al. Repeated adeno-
associated virus serotype 2 aerosol-mediated cystic fibrosis transmembrane regulator
gene transfer to the lungs of patients with cystic fibrosis: a multicenter, double-blind,
placebo-controlled trial. Chest 2004; 125: 509–521.
17 Koehler DR, Frndova H, Leung K, Louca E, Palmer D, Ng P et al. Aerosol delivery of an
enhanced helper-dependent adenovirus formulation to rabbit lung using an intra-
tracheal catheter. J Gene Med 2005; 7: 1409–1420.
18 Koping-Hoggard M, Issa MM, Kohler T, Tronde A, Varum KM, Artursson P. A minia-
turized nebulization catheter for improved gene delivery to the mouse lung. J Gene Med
2005; 7: 1215–1222.
19 Arnon TI, Achdout H, Lieberman N, Gazit R, Gonen-Gross T, Katz G et al. The
mechanisms controlling the recognition of tumor- and virus-infected cells by NKp46.
Blood 2004; 103: 664–672.
20 Komaru A, Ueda Y, Furuya A, Tanaka S, Yoshida K, Kato T et al. Sustained and
NK/CD4+ Tcell-dependent efficient prevention of lung metastasis induced by dendritic
cells harboring recombinant Sendai virus. J Immunol 2009; 183: 4211–4219.
21 Limberis MP, Wilson JM. Adeno-associated virus serotype 9 vectors transduce murine
alveolar and nasal epithelia and can be readministered. Proc Natl Acad Sci USA
2006; 103: 12993–12998.
22 Pichla-Gollon SL, Lin SW, Hensley SE, Lasaro MO, Herkenhoff-Haut L, Drinker M et al.
Effect of preexisting immunity on an adenovirus vaccine vector: in vitro neutralization
assays fail to predict inhibition by antiviral antibody in vivo. J Virol 2009; 83:
23 Kobayashi M, Iida A, Ueda Y, Hasegawa M. Pseudotyped lentivirus vectors derived
from simian immunodeficiency virus SIVagm with envelope glycoproteins from
paramyxovirus. J Virol 2003; 77: 2607–2614.
24 Mitomo K, Griesenbach U, Inoue M, Somerton L, Meng C, Akiba E et al. Toward
Gene Therapy for Cystic Fibrosis Using a Lentivirus Pseudotyped With Sendai Virus
Envelopes. Mol Ther 2010; 18: 1173–1182.
25 Hirata T, Iida A, Shiraki-Iida T, Kitazato K, Kato A, Nagai Y et al. An improved method
for recovery of F-defective Sendai virus expressing foreign genes from cloned cDNA.
J Virol Methods 2002; 104: 125–133.
26 Lee ER, Marshall J, Siegel CS, Jiang C, Yew NS, Nichols MR et al. Detailed analysis of
structures and formulations of cationic lipids for efficient gene transfer to the lung.
Hum Gene Ther 1996; 7: 1701–1717.
27 McLachlan G, Davidson DJ, Stevenson BJ, Dickinson P, Vidson-Smith H, Dorin JR et al.
Evaluation in vitro and in vivo of cationic liposome-expression construct complexes for
cystic fibrosis gene therapy. Gene Therapy 1995; 2: 614–622.
Validation of recombinant Sendai virus in sheep
U Griesenbach et al